写一个达梦数据库函数，传入一个要生成的表名(以_年份结束）和要生成的年份和要复制的表名(以_年份结束），复制一个年表生成新的年表，表名结尾以_年份结束，先判断要生成的表是否存在，存在则不生成新表，不存在则生成新表，要复制字段信息、字段注释、唯一约束、表注释、主键、索引、分区等信息 要防止索引名冲突，分区按照要复制的表的规则生成

以下是一个基于达梦数据库的函数，实现了根据传入的参数生成新表的功能：

CREATE FUNCTION generate_new_table(table_name VARCHAR(50), year INT, source_table_name VARCHAR(50))
RETURNS VARCHAR(50)
BEGIN
    DECLARE new_table_name VARCHAR(50);
    DECLARE source_table_id INT;
    DECLARE index_prefix VARCHAR(20);
    DECLARE partition_clause VARCHAR(100);
    DECLARE source_table_comment VARCHAR(200);
    DECLARE source_table_partition_clause VARCHAR(100);
    DECLARE column_name VARCHAR(50);
    DECLARE column_type VARCHAR(50);
    DECLARE column_comment VARCHAR(200);
    DECLARE column_is_unique INT;
    DECLARE column_is_primary INT;
    DECLARE column_is_indexed INT;
    DECLARE column_is_partition INT;
    DECLARE column_order INT;
    DECLARE index_name VARCHAR(50);
    DECLARE index_type VARCHAR(20);
    DECLARE index_columns VARCHAR(200);
    DECLARE partition_column VARCHAR(50);
    DECLARE partition_value VARCHAR(50);
    DECLARE partition_start_date DATE;
    DECLARE partition_end_date DATE;
    DECLARE partition_interval INT;
    DECLARE partition_num INT;
    
    SET new_table_name = CONCAT(table_name, '_', year);
    SELECT COUNT(*) INTO source_table_id FROM DM_TABLE WHERE TABLE_NAME = source_table_name;
    IF source_table_id = 0 THEN
        RETURN 'Source table does not exist.';
    END IF;
    SELECT INDEX_PREFIX INTO index_prefix FROM DM_INDEX WHERE TABLE_NAME = source_table_name;
    SELECT PARTITION_CLAUSE INTO source_table_partition_clause FROM DM_TABLE WHERE TABLE_NAME = source_table_name;
    SELECT TABLE_COMMENT INTO source_table_comment FROM DM_TABLE WHERE TABLE_NAME = source_table_name;
    
    IF EXISTS (SELECT * FROM DM_TABLE WHERE TABLE_NAME = new_table_name) THEN
        RETURN 'Table already exists.';
    END IF;
    
    CREATE TABLE new_table_name (
        -- Copy columns from source table
        SELECT COLUMN_NAME, COLUMN_TYPE, COLUMN_COMMENT, COLUMN_IS_UNIQUE, COLUMN_IS_PRIMARY, COLUMN_IS_INDEXED, COLUMN_IS_PARTITION, COLUMN_ORDER 
        FROM DM_COLUMNS
        WHERE TABLE_NAME = source_table_name
        INTO column_name, column_type, column_comment, column_is_unique, column_is_primary, column_is_indexed, column_is_partition, column_order
    );
    
    -- Add table comment
    ALTER TABLE new_table_name COMMENT source_table_comment;
    
    -- Add unique constraints
    SELECT CONCAT('ALTER TABLE ', new_table_name, ' ADD CONSTRAINT ', CONSTRAINT_NAME, ' UNIQUE (', COLUMN_NAME, ');')
    FROM DM_CONSTRAINTS
    WHERE TABLE_NAME = source_table_name AND CONSTRAINT_TYPE = 'U'
    INTO @unique_constraint_sql;
    PREPARE stmt FROM @unique_constraint_sql;
    EXECUTE stmt;
    
    -- Add primary key
    SELECT GROUP_CONCAT(COLUMN_NAME ORDER BY KEY_SEQ SEPARATOR ', ') INTO index_columns
    FROM DM_INDEX_COLUMNS
    WHERE TABLE_NAME = source_table_name AND INDEX_NAME = 'PRIMARY';
    IF index_columns IS NOT NULL THEN
        ALTER TABLE new_table_name ADD PRIMARY KEY (index_columns);
    END IF;
    
    -- Add indexes
    SELECT INDEX_NAME, INDEX_TYPE, GROUP_CONCAT(COLUMN_NAME ORDER BY COLUMN_ORDER SEPARATOR ', ') INTO index_name, index_type, index_columns
    FROM DM_INDEX_COLUMNS
    WHERE TABLE_NAME = source_table_name AND INDEX_NAME <> 'PRIMARY'
    GROUP BY INDEX_NAME, INDEX_TYPE;
    IF index_name IS NOT NULL THEN
        SET index_name = CONCAT(index_prefix, '_', index_name);
        SET @index_sql = CONCAT('CREATE ', INDEX_TYPE, ' INDEX ', index_name, ' ON ', new_table_name, ' (', index_columns, ');');
        PREPARE stmt FROM @index_sql;
        EXECUTE stmt;
    END IF;
    
    -- Add partitions
    SELECT PARTITION_COLUMN, PARTITION_START_DATE, PARTITION_END_DATE, PARTITION_INTERVAL, PARTITION_NUM
    FROM DM_PARTITION
    WHERE TABLE_NAME = source_table_name
    INTO partition_column, partition_start_date, partition_end_date, partition_interval, partition_num;
    IF partition_column IS NOT NULL THEN
        SET partition_clause = CONCAT('PARTITION BY RANGE (', partition_column, ') (');
        WHILE partition_start_date <= partition_end_date DO
            SET partition_value = DATE_FORMAT(partition_start_date, '%Y-%m-%d');
            SET partition_clause = CONCAT(partition_clause, '\n', 'PARTITION p', partition_order, ' VALUES LESS THAN (\'', partition_value, '\') ENGINE = InnoDB,');
            SET partition_start_date = ADDDATE(partition_start_date, INTERVAL partition_interval DAY);
            SET partition_order = partition_order + 1;
            IF partition_order > partition_num THEN
                LEAVE;
            END IF;
        END WHILE;
        SET partition_clause = SUBSTRING(partition_clause, 1, LENGTH(partition_clause) - 1);
        SET partition_clause = CONCAT(partition_clause, '\n', ')', source_table_partition_clause);
        ALTER TABLE new_table_name PARTITION BY RANGE (partition_column) (
            EXECUTE IMMEDIATE partition_clause
        );
    END IF;
    
    RETURN new_table_name;
END;

这个函数的逻辑如下：

1. 首先检查要生成的表是否已经存在，如果存在则返回错误信息。
2. 检查要复制的表是否存在，如果不存在则返回错误信息。
3. 根据要复制的表的信息，复制字段信息、字段注释、唯一约束、表注释、主键、索引、分区等信息，生成新表。
4. 防止索引名冲突，为新表的索引名添加前缀。
5. 根据要复制的表的分区信息，生成新表的分区信息。

这个函数使用了多个DM_*系统表来获取要复制的表的信息，包括DM_TABLE、DM_COLUMNS、DM_CONSTRAINTS、DM_INDEX、DM_INDEX_COLUMNS和DM_PARTITION等。在复制索引和分区时，使用了动态SQL来执行ALTER TABLE和CREATE INDEX语句，这样可以避免索引名和分区名冲突的问题。